I-vector based speaker recognition using advanced channel compensation techniques

This paper investigates advanced channel compensation techniques for the purpose of improving i-vector speaker verification performance in the presence of high intersession variability using the NIST 2008 and 2010 SRE corpora. The performance of four channel compensation techniques: (a) weighted maximum margin criterion (WMMC), (b) source-normalized WMMC (SN-WMMC), (c) weighted linear discriminant analysis (WLDA), and; (d) source-normalized WLDA (SN-WLDA) have been investigated. We show that, by extracting the discriminatory information between pairs of speakers as well as capturing the source variation information in the development i-vector space, the SN-WLDA based cosine similarity scoring (CSS) i-vector system is shown to provide over 20% improvement in EER for NIST 2008 interview and microphone verification and over 10% improvement in EER for NIST 2008 telephone verification, when compared to SN-LDA based CSS i-vector system. Further, score-level fusion techniques are analyzed to combine the best channel compensation approaches, to provide over 8% improvement in DCF over the best single approach, (SN-WLDA), for NIST 2008 interview/ telephone enrolment-verification condition. Finally, we demonstrate that the improvements found in the context of CSS also generalize to state-of-the-art GPLDA with up to 14% relative improvement in EER for NIST SRE 2010 interview and microphone verification and over 7% relative improvement in EER for NIST SRE 2010 telephone verification.

Designing Innovative Business Models : A Design Catalyst Learning Resource

The term Design Led Innovation is emerging as a fundamental business process, which is rapidly being adopted by large as well as small to medium sized firms. The value that design brings to an organisation is a different way of thinking, of framing situations and possibilities, doing things and tackling problems: essentially a cultural transformation of the way the firm undertakes its business. Being Design Led is increasingly being seen by business as a driver of company growth, allowing firms to provide a strong point of difference to its stakeholders. Achieving this Design Led process, requires strong leadership to enable the organisation to develop a clear vision for top line growth. Specifically, based on deep customer insights and expanded through customer and stakeholder engagements, the outcomes of which are then adopted by all aspects of the business. To achieve this goal, several tools and processes are available, which need to be linked to new organisational capabilities within a business transformation context. The Design Led Innovation Team focuses on embedding tools and processes within an organisation and matching this with design leadership qualities to enable companies to create breakthrough innovation and achieve sustained growth, through ultimately transforming their business model. As all information for these case studies was derived from publicly accessed data, this resource is not intended to be used as reference material, but rather is a learning tool for designers to begin to consider and explore businesses at a strategic level. It is not the results that are key, but rather the process and philosophies that were used to create these case studies and disseminate this way of thinking amongst the design community. It is this process of unpacking a business guided by the framework of Osterwalder’s Business Model Canvas* which provides an important tool for designers to gain a greater perspective of a company’s true innovation potential.

Linking Karumba : Creating Sustainable Connections

Linking Karumba: Creating Sustainable Connections This exhibition showcases the work of 3rd -4th year undergraduate landscape architecture, architecture, Industrial Design, Environmental Engineering, Civil Engineering students in response to issues of sustainability in the Gulf of Carpentaria town of Karumba. It presented the work to the Karumba and Carpentaria Shire community. 16 students and four staff set off on a 2488km journey to undertake the first half of the Carpentaria Project: a fortnight-long strategic planning project entitled Linking Karumba to encourage social, economic, environmental and cultural linkages across the town. Karumba, along with the nearby town of Normanton, is one of Queensland’s most remote settlements. Its economy is based on fishing, tourism, and mining. It has two centres, 2.5km apart by river, or 9km by road. This physical disconnect was identified by Carpentaria Shire Council (CSC) and the Karumba Progress Association (KPA) as a source of socio-cultural disconnection, which formed the basis of our project brief. Student designs were highly responsive to the character of Karumba’s culture and environment, indicating remarkable levels of immersion, and attracting $830 000 in Qld. state government funding for implementation. The Exhibition Four groups of four students produced four strategic planning and design options toward this future: Make the Switch: Alice Anonuevo, Michael Marriott, Carla Priestley & Grant Harvey Realigning the Systems: Claudia Bergs, Rebecca Stephens, Anna Coulson & Lois Kerrigan Diversification of Experience: Rebecca North, Kyle Bush, Debra Sullivan & Jenna Green The River is the Main Street: Ashley Nicholson, Monica Kuiken, Dean Bowen & Bill Schild

QUT Linking Karumba Project

QUT Linking Karumba Project This exhibition showcases the work of 3rd -4th year undergraduate landscape architecture, architecture, Industrial Design, Environmental Engineering, Civil Engineering students in response to issues of sustainability in the Gulf of Carpentaria town of Karumba. It presented the final, polished set of work to the Karumba and Carpentaria Shire community, following revisions in line with feedback from the 2008 exhibition. 16 students and four staff set off on a 2488km journey to undertake the first half of the Carpentaria Project: a fortnight-long strategic planning project entitled Linking Karumba to encourage social, economic, environmental and cultural linkages across the town. Karumba, along with the nearby town of Normanton, is one of Queensland’s most remote settlements. Its economy is based on fishing, tourism, and mining. It has two centres, 2.5km apart by river, or 9km by road. This physical disconnect was identified by Carpentaria Shire Council (CSC) and the Karumba Progress Association (KPA) as a source of socio-cultural disconnection, which formed the basis of our project brief. Student designs were highly responsive to the character of Karumba’s culture and environment, indicating remarkable levels of immersion, and attracting $830 000 in Qld. state government funding for implementation. The Exhibition Four groups of four students produced four strategic planning and design options toward this future: Make the Switch: Alice Anonuevo, Michael Marriott, Carla Priestley & Grant Harvey Realigning the Systems: Claudia Bergs, Rebecca Stephens, Anna Coulson & Lois Kerrigan Diversification of Experience: Rebecca North, Kyle Bush, Debra Sullivan & Jenna Green The River is the Main Street: Ashley Nicholson, Monica Kuiken, Dean Bowen & Bill Schild

Developing a method for assessing product inclusivity

In order to develop more inclusive products and services, designers need a means of assessing the inclusivity of existing products and new concepts. Following previous research on the development of scales for inclusive design at University of Cambridge, Engineering Design Centre (EDC) [1], this paper presents the latest version of the exclusion audit method. For a specific product interaction, this estimates the proportion of the Great British population who would be excluded from using a product or service, due to the demands the product places on key user capabilities. A critical part of the method involves rating of the level of demand placed by a task on a range of key user capabilities, so the procedure to perform this assessment was operationalised and then its reliability was tested with 31 participants. There was no evidence that participants rated the same demands consistently. The qualitative results from the experiment suggest that the consistency of participants’ demand level ratings could be significantly improved if the audit materials and their instructions better guided the participant through the judgement process.

Repeating patterns : strategies to assist young students to generalise the mathematical structure

This paper focuses on very young students' ability to engage in repeating pattern tasks and identifying strategies that assist them to ascertain the structure of the pattern. It describes results of a study which is part of the Early Years Generalising Project (EYGP) and involves Australian students in Years 1 to 4 (ages 5-10). This paper reports on the results from the early years' cohort (Year 1 and 2 students). Clinical interviews were used to collect data concerning students' ability to determine elements in different positions when two units of a repeating pattern were shown. This meant that students were required to identify the multiplicative structure of the pattern. Results indicate there are particular strategies that assist students to predict these elements, and there appears to be a hierarchy of pattern activities that help students to understand the structure of repeating patterns.

A UKF-based estimation strategy for actuator fault detection of UASs

This paper presents a recursive strategy for online detection of actuator faults on a unmanned aerial system (UAS) subjected to accidental actuator faults. The proposed detection algorithm aims to provide a UAS with the capability of identifying and determining characteristics of actuator faults, offering necessary flight information for the design of fault-tolerant mechanism to compensate for the resultant side-effect when faults occur. The proposed fault detection strategy consists of a bank of unscented Kalman filters (UKFs) with each one detecting a specific type of actuator faults and estimating correspond- ing velocity and attitude information. Performance of the proposed method is evaluated using a typical nonlinear UAS model and it is demonstrated in simulations that our method is able to detect representative faults with a sufficient accuracy and acceptable time delay, and can be applied to the design of fault-tolerant flight control systems of UASs.

Online calibration of stereo rigs for long-term autonomy

Stereo-based visual odometry algorithms are heavily dependent on an accurate calibration of the rigidly fixed stereo pair. Even small shifts in the rigid transform between the cameras can impact on feature matching and 3D scene triangulation, adversely affecting pose estimates and applications dependent on long-term autonomy. In many field-based scenarios where vibration, knocks and pressure change affect a robotic vehicle, maintaining an accurate stereo calibration cannot be guaranteed over long periods. This paper presents a novel method of recalibrating overlapping stereo camera rigs from online visual data while simultaneously providing an up-to-date and up-to-scale pose estimate. The proposed technique implements a novel form of partitioned bundle adjustment that explicitly includes the homogeneous transform between a stereo camera pair to generate an optimal calibration. Pose estimates are computed in parallel to the calibration, providing online recalibration which seamlessly integrates into a stereo visual odometry framework. We present results demonstrating accurate performance of the algorithm on both simulated scenarios and real data gathered from a wide-baseline stereo pair on a ground vehicle traversing urban roads.

Legal activism for ensuring environmental justice

This article reviews some of the roles environmental lawyers have played in ensuring environmental justice in Bangladesh. It leans on law and social movement theories to explicate the choice (and ensuing success) of litigation as a movement strategy in Bangladesh. The activists successfully moved the courts to read the right to a decent environment into the fundamental right to life, and this has had the far-reaching effect of constituting a basis for standing for the activists and other civil society organisations. The activists have also sought to introduce emerging international law principles into the jurisprudence of the courts. These achievements notwithstanding, the paper notes that litigation is not a sustainable way to institute enduring environmental protection in any jurisdiction and recommends the utilisation of the reputation and recognition gained through litigation to deploy or encourage more sustainable strategies.

High altitude stereo visual odometry

Stereo visual odometry has received little investigation in high altitude applications due to the generally poor performance of rigid stereo rigs at extremely small baseline-to-depth ratios. Without additional sensing, metric scale is considered lost and odometry is seen as effective only for monocular perspectives. This paper presents a novel modification to stereo based visual odometry that allows accurate, metric pose estimation from high altitudes, even in the presence of poor calibration and without additional sensor inputs. By relaxing the (typically fixed) stereo transform during bundle adjustment and reducing the dependence on the fixed geometry for triangulation, metrically scaled visual odometry can be obtained in situations where high altitude and structural deformation from vibration would cause traditional algorithms to fail. This is achieved through the use of a novel constrained bundle adjustment routine and accurately scaled pose initializer. We present visual odometry results demonstrating the technique on a short-baseline stereo pair inside a fixed-wing UAV flying at significant height (~30-100m).

Robust scale initialization for long-range stereo visual odometry

Achieving a robust, accurately scaled pose estimate in long-range stereo presents significant challenges. For large scene depths, triangulation from a single stereo pair is inadequate and noisy. Additionally, vibration and flexible rigs in airborne applications mean accurate calibrations are often compromised. This paper presents a technique for accurately initializing a long-range stereo VO algorithm at large scene depth, with accurate scale, without explicitly computing structure from rigidly fixed camera pairs. By performing a monocular pose estimate over a window of frames from a single camera, followed by adding the secondary camera frames in a modified bundle adjustment, an accurate, metrically scaled pose estimate can be found. To achieve this the scale of the stereo pair is included in the optimization as an additional parameter. Results are presented both on simulated and field gathered data from a fixed-wing UAV flying at significant altitude, where the epipolar geometry is inaccurate due to structural deformation and triangulation from a single pair is insufficient. Comparisons are made with more conventional VO techniques where the scale is not explicitly optimized, and demonstrated over repeated trials to indicate robustness.

Valuing student and community voices in the university : action research as a framework for community service-learning

The Community Service-learning Lab (the Lab) was initiated as a university-wide service-learning experience at an Australian university. The Lab engages students, academics, and key community organisations in interdisciplinary action research projects to support student learning and to explore complex and ongoing problems nominated by the community partners. The current study uses feedback from the first offering of the Lab and focuses on exploring student experiences of the service learning project using an action research framework. Student reflections on this experience have revealed some positive outcomes of the Lab such as an appreciation for positive and strengths-based change. These outcomes are corroborated by collected reflections from community partners and academics. The students also identified challenges balancing the requirements for assessment and their goals to serve the community partner’s needs. This feedback has provided vital information for the academic team, highlighting the difficulties in balancing the agenda of the academic framework and the desire to give students authentic experiences.

Sun, sea, sex, sand and conservatism in the Australian television soap opera

A recent issue of Young People Now (November 1995) mentioned the new (UK) television soap opera Hollyoaks by Phil Redmond, which raises the issue of the role of ‘soap operas’ (hereafter referred to as soaps) in the daily lives of young people. The term ‘soap’ originates with the sponsorship of radio and television programmes by companies such as Proctor and Gamble who in America in 1932 used a daytime radio domestic comedy, The Puddle Family to advertise Oxydol, a washing powder. The first British television soap was The Grove Family (BBC 1954-7) was followed by Emergency Ward Ten (ATV 1957-67), Coronation Street (Granada Television 1960-present) and Eastenders (BBC 1985-present). Australian soaps are especially popular in Britain and of potential interest to those who work with young people, because they have a high proportion of youthful looking actors and actresses and frequently depict scenes involving young people and apparent ‘real’ teenage dilemmas. On one level it may be commendable that actors who are young(ish) somewhere between the ages of 14 and 25 play roles that are ostensibly about young people and their alleged problems. However, the casting of young, largely unknown, actors reflects more the political economy of soaps in their relative cheapness and dispensability, rather than any genuine attempt to create an oppositional text for, about and by young people (Paterson 1986).

Visual front-end wars : Viola-Jones face detector vs Fourier Lucas-Kanade

The performance of visual speech recognition (VSR) systems are significantly influenced by the accuracy of the visual front-end. The current state-of-the-art VSR systems use off-the-shelf face detectors such as Viola- Jones (VJ) which has limited reliability for changes in illumination and head poses. For a VSR system to perform well under these conditions, an accurate visual front end is required. This is an important problem to be solved in many practical implementations of audio visual speech recognition systems, for example in automotive environments for an efficient human-vehicle computer interface. In this paper, we re-examine the current state-of-the-art VSR by comparing off-the-shelf face detectors with the recently developed Fourier Lucas-Kanade (FLK) image alignment technique. A variety of image alignment and visual speech recognition experiments are performed on a clean dataset as well as with a challenging automotive audio-visual speech dataset. Our results indicate that the FLK image alignment technique can significantly outperform off-the shelf face detectors, but requires frequent fine-tuning.

Nonlinear actuator fault detection for small-scale UAS

This paper presents a recursive strategy for online detection of actuator faults on a unmanned aerial system (UAS) subjected to accidental actuator faults. The proposed detection algorithm aims to provide a UAS with the capability of identifying and determining characteristics of actuator faults, offering necessary flight information for the design of fault-tolerant mechanism to compensate for the resultant side-effect when faults occur. The proposed fault detection strategy consists of a bank of unscented Kalman filters (UKFs) with each one detecting a specific type of actuator faults and estimating corresponding velocity and attitude information. Performance of the proposed method is evaluated using a typical nonlinear UAS model and it is demonstrated in simulations that our method is able to detect representative faults with a sufficient accuracy and acceptable time delay, and can be applied to the design of fault-tolerant flight control systems of UASs.